Computational modeling of myocardial contractility to personalize the response to pharmacological treatment in heart failure with reduced ejection fraction

Heart Failure (HF) is a complex clinical syndrome characterized by impaired cardiac function, which can manifest across a wide spectrum of severity and underlying mechanisms. Computational modeling offers a powerful approach to study cardiac dynamics and assess therapeutic strategies across different stages of HF. Selective cardiac myosin activators can enhance contractility without increasing oxygen consumption or inducing arrhythmias. Although showing efficacy in patients with HF with reduced ejection fraction (HFrEF) [2,3], its clinical benefit varies significantly across subgroups [1].

This study presents a comprehensive computational cardiac model capable of simulating electromechanical behavior in healthy and failing hearts. The model is designed to evaluate the impact of pharmacological interventions under a range of hemodynamic and structural conditions representative of various HF phenotypes. Computational simulations are used to identify potential drug candidates that are likely to have the desired effects on contractility and predict potential side effects.

An integrated biventricular hemo-electromechanical model has been developed, based on a multiscale approach that combines: a modified cellular model of myosin activation, which can incorporate drug-specific kinetics; and a three-dimensional model of myocardial dynamics calibrated with experimental and clinical data [4]. The model enables the simulation of systolic duration, sarcomere-level active tension generation, and cardiac output across virtual patients with varying HF severity. Simulations include both baseline and drug-modulated conditions, supporting the study of differential responses across a virtual patient cohort.

Preliminary results show that the model can predict the increase in stroke volume with myosin activators and the improvement in left ventricular ejection fraction (LVEF) in cases of severe HF. Using the model in the virtual population will allow stratification of patients with the most favorable response to the treatment. Initial simulations demonstrate the model’s ability to reproduce key features of cardiac performance across a spectrum of HF severity. Figure 1 shows PVloops of a healthy virtual patient(h), with induced HF (hf) and after myosin activator pharmacological treatment. The model predicts individualized changes in stroke volume and LVEF, consistent with clinically observed heterogeneity in treatment response.

This computational framework provides a robust and flexible platform for studying cardiac function under healthy and pathological conditions. It enables patient-specific assessment of therapeutic strategies, including the evaluation of drug efficacy and optimization of treatment selection. The model holds promise for supporting translational research and guiding clinical decision-making, pending further validation through prospective studies.Figure 1